OS Manuals

/usr/bin/onnxruntime_test

Cross-platform, high performance ML inferencing and training accelerator

From py3.12-onnxruntime 1.31.1-r0 (enterprise-packages), captured 2026-11-08 by running onnxruntime_test --help (exit 0). permalink

usage: onnxruntime_test [-h] [--debug] [--profile] [--symbolic_dims SYMBOLIC_DIMS]
                        model_path [num_iters]

Simple ONNX Runtime Test Tool.

positional arguments:
  model_path            model path
  num_iters             model run iterations. default=1000

options:
  -h, --help            show this help message and exit
  --debug               pause execution to allow attaching a debugger.
  --profile             enable chrome timeline trace profiling.
  --symbolic_dims SYMBOLIC_DIMS
                        Comma separated name=value pairs for any symbolic dimensions in the model
                        input. e.g. --symbolic_dims batch=1,seqlen=5. If not provided, the value
                        of 1 will be used for all symbolic dimensions.
Options (4)
OptionDescriptionSource
-h, --helpshow this help message and exithelp
--debugpause execution to allow attaching a debugger.help
--profileenable chrome timeline trace profiling.help
--symbolic_dims SYMBOLIC_DIMSComma separated name=value pairs for any symbolic dimensions in the model input.help

Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.